News
What the people building AI are posting.
Straight from X: the labs, the researchers and the builders worth following, in one place and in the order they posted.
California Gov. Gavin Newsom signed an executive order giving an expert panel two months to recommend tougher AI safety laws. On the table: a "kill switch," outside monitors inside frontier labs, and required safety plans. Newsom says the industry is "begging for regulation."
View on X
GPT-6 Astra’s “techno-utopia,” with our logo snuck in. Animation and soundtrack created in JavaScript:
View on XTop stories in tech today: - Meta’s next smart glasses could lose the camera - OpenAI is footing the bill for more biology data - Scientists grow a human cortex inside mice - GLP-1s are finding a life beyond weight loss
View on X
Top stories in AI today: - OpenAI’s new rules for reporting model misbehavior - Rowan’s Corner: Why I keep cancelling great AI tools - Test AI video object swaps with Higgsfield - GPT-6 Astra helps crack unsolved WWII message
View on X
World Labs' new Atlas model turns a single photograph into a 3D world you can navigate. It can also turn ordinary phone footage into simulated spaces for training robots:
View on XGPT-6 Astra is doing WWII codebreaking now. Carter Leffen said the model cracked a German Army radio message from 1941 that had not yet been decoded: "Please specify the route of march. I am in Rosenow, Rosenow. Immediate reply by radio." The run took ~10 hours and 650M tokens on a /goal of "don't stop working until you solve the problem".
View on XOpenAI is testing ads you can talk to. In its demo, a grocery ad under a ChatGPT dinner answer opens a separate chat with the store's sponsored AI agent, which suggests recipes and points you to the ingredients. Also new for ChatGPT ads: - Ads that adapt their wording to match the conversation and user's language - Run ChatGPT ads from inside Shopify and HubSpot - Ad campaigns built + managed via prompting ChatGPT - AI-suggested ad copy and images Currently in testing with select U.S. advertisers
View on XTop stories in robotics today: - Agility’s Digit crouches to keep humans safe - Waymo to launch robotaxis in Tokyo - How Unitree drove humanoid prices so low - UBTECH’s factory can build 10K humanoids a year
View on X
Top stories in AI today: - Zuckerberg breaks ranks on the AI slowdown - Nate's Notebook: Same-hole thinking - A beginner’s guide to ChatGPT Work - Google DeepMind starts an AGI think tank
View on X
The Cowork / Chat split (and ditto for ChatGPT Work) has always felt like an awkward divide, without a crystal clear reason for users to live in one over the other. Anthropic ripping the bandaid on merging the two: x.com/claudeai/statu…
View on XGoogle DeepMind just launched the DeepMind Institute, an in-house AGI think tank run by Demis Hassabis, James Manyika, and Shane Legg. The Institute released several essays on day one, including one weighing policies for AI-driven job loss. "While AI can still sometimes fail at basic tasks and lacks the consistency and creativity to meet the bar of full AGI, we expect those gaps to be closed soon."
View on X
Posts appear as published on X and link to the original. Replies and reposts are left out, and anything older than 14 days drops off.